基于轨道空间压缩的混沌神经网络控制
Controlling Chaos in a Neural Network Based on the Orbit Space Compression
-
摘要: 该文提出了基于轨道空间压缩的混沌神经网络控制方法,利用该方法对混沌神经网络进行控 制,使神经网络的输出稳定地收敛于与网络起始模式有最小汉明距离的存储模式或其反相模式上。该控制方法简单易行,物理意义明确。
-
关键词:
- 混沌控制; 混沌神经网络; 轨道空间压缩
Abstract: In this paper, a controlling chaos method of the orbit space compression is proposed for a Chaotic Neural Network(CNN). The computer simulation of the chaotic behaviors of the CNN proves that each pattern can be controlled using the orbit space compression. Starting from any initial state the CNN can converge in a stored pattern or its inverse pattern, which has the smallest Hamming distance with the initial state. The controlling method of the orbit space compression shows clear physical meaning and can be easily carried out. -
曹志彤.混沌神经网络的动态联想记忆.浙江大学学报(自然科学版),1998,增刊:330-335.[2]Aihara K, Takabe T, Toyoda M. Chaotic neural networks[J].Phys. Lett.1990, 144(6,7):333-340[3]Adachi M, Aihara K. Associative dynamics in a chaotic neural network[J].Neural Networks.1997,10(1):83-98[4]Kobori E R, Ikoda K, Nakayama K. A model of dynamic associative memory. IEEE International Conference on Neural Networks Conference Proceedings, Washington, DC, USA, 1996, 2: 804-809.[5]何国光,曹志彤.混沌神经网络的信息搜素.物理学报,2001,50(11):2103-2107.[6]Ott E, Grebogi C, Yorke J A. Controlling chaos[J].Phys. Rev. Lett.1990, 64:1196-1199[7]罗晓曙.利用相空间压缩实现混沌与超混沌控制.物理学报,1999,48(3):402-407.
计量
- 文章访问数: 2297
- HTML全文浏览量: 116
- PDF下载量: 556
- 被引次数: 0